ROI, ROAS, and ROMI — the definitive version
These three get used as synonyms in marketing conversations and they are not interchangeable. The distinction is worth getting right because it is the most common source of a marketing team and a finance team reaching opposite conclusions from the same campaign.
- ROAS = revenue ÷ ad spend. A multiple. Media costs only, revenue not profit. Answers “is this ad account working?” Calculate ROAS.
- Marketing ROI = (return − cost) ÷ cost. A percentage. All marketing costs, ideally measured in profit. Answers “is marketing worth what we put into it?”
- ROMI = the same formula applied to the whole marketing programme rather than a single campaign. Answers the same question at budget level.
The practical consequence: a campaign at 4x ROAS can be ROI-negative. $100,000 of media returning $400,000 of revenue looks excellent until you add $80,000 of salaries and tooling and apply a 40% gross margin — at which point $160,000 of gross profit against $180,000 of cost is a loss.
What belongs in the investment
The cost side is where most marketing ROI figures quietly fall apart. If the number only contains media spend, it is a ROAS wearing a percentage sign. A defensible marketing investment includes salaries and contractor costs for everyone doing the work, agency retainers, the marketing software stack, creative and production, and any discounting or incentive used to close the sale.
One genuine judgement call: how to treat spend on assets that keep producing. Content and organic search have production costs today and returns for years. Expensing them entirely against the current period understates ROI; amortising them over an assumed life is more accurate and harder to defend. Pick one, document it, and apply it consistently.
Attributed revenue versus incremental revenue
Attribution tells you which touchpoints a converting customer encountered. It does not tell you whether they would have bought anyway. Branded search is the clearest case: it converts superbly and a meaningful share of that revenue would have arrived through an organic result at no cost.
Incrementality — measured with geo holdouts, matched-market tests, or simply pausing a channel and watching total revenue — is the only way to separate the two. It is more work than reading a dashboard, and it is the difference between an ROI figure you can defend in a board meeting and one that collapses under the first hard question.
Time horizons change the answer
Marketing ROI measured in a single month systematically favours whatever converts fastest. Retargeting and branded search will always win that comparison; brand building and content will always lose it, because their returns land outside the measurement window.
Run the calculation over a period at least as long as your sales cycle, and prefer a rolling twelve months for anything involving organic search or content. The ranking of channels frequently reverses when you widen the window, which is precisely the information the exercise is meant to produce.
Raising marketing ROI
Both sides of the fraction are available, and the cost side is usually the neglected one. On returns: raise close rate and average order value, and improve retention so each acquired customer contributes more — LTV:CAC is the diagnostic there. On costs: replace per-unit media cost with owned channels, and automate the manual work that consumes salary without producing output. The automation savings calculator prices that second one directly.